This paper is meant to elimiante te assumption that written text is fixed, using a handwriting texture analysis. the results are 96.0% accurate. signiture identification is a very challenging and active research topic. hanwriting analysis provides man different ways of identifying the writer of a piece of handwriting. this process is used in order to verify the identity of a person for what ever the reason. It requires the writer to write
the same fixed text that had been previously collected.
1. What is the main idea of the paper? This article is about identifying people based on their handwriting.
2. What experiment, if any, was performed?
The texture analysis has been tested on a set of 800 handwriting samples. They used twenty writers and took about fourty samples per writer. Examples of handwriting by thesepeople are shown in Fig. 3. For the purpose of the classification experiments 25 non-overlapping handwriting blocks were extracted from each person. Each sample was selected from an A4 page. the sample images were divided into 15 test images per writer followed by 15 training and 10 test images. In this paper they use the Gabor filters ,WED classifiers, and the K-NN classifier. These filters have been proven to be successful in extracting similar features from handwriting samples
3. Was the experiment ) GS or BS? That is, was there a large sample size, a cause-and-effect relationship, etc. If they had more test subjects this would be better science but with only 20 people for the test subjests it is bad science.
4. What did the authors determine about forensic graphology? The Results achieved were very promising,and an identification accuracy of 96.0% was obtained using the WED classifier. The K-NN classifiergave somewhat poor results. The author believes that Graphology is a good thing to use as evidence and that no two hand writing samples were exaclt ythe same. Graphology can be used to identify a person within a 96.0% accuracy.
5. Name at least five interesting facts from the paper that you think the rest of the class might appreciate.
Texture analysis cannot be applied directly to handwriting images, as texture is affected by different word spacing, varying line spacing, etc.
The handwriting may contain lines of different point size and di!erent spacing between lines, words and characters
In principle, any texture analysis technique can be applied to extract features from each uniform block of handwriting.
In this paper we use frequencies.
The whole analysis process is done with mathmatics.
The relative sample invariance can be calculated by means of the ratio of the standard deviation of writer sample features to the mean of those sample features.
PERSONAL IDENTIFICATION BASED ON HANDWRITING:
This paper is meant to elimiante te assumption that written text is fixed, using a handwriting texture analysis. the results are 96.0% accurate. signiture identification is a very challenging and active research topic. hanwriting analysis provides man different ways of identifying the writer of a piece of handwriting. this process is used in order to verify the identity of a person for what ever the reason. It requires the writer to writethe same fixed text that had been previously collected.
1. What is the main idea of the paper?
This article is about identifying people based on their handwriting.
2. What experiment, if any, was performed?
The texture analysis has been tested on a set of 800 handwriting samples. They used twenty writers and took about fourty samples per writer. Examples of handwriting by these people are shown in Fig. 3.
For the purpose of the classification experiments 25 non-overlapping handwriting blocks were extracted from each person. Each sample was selected from an A4 page. the sample images were divided into 15 test images per writer followed by 15 training and 10 test images. In this paper they use the Gabor filters ,WED classifiers, and the K-NN classifier. These filters have been proven to be successful in extracting similar features from handwriting samples
3. Was the experiment ) GS or BS? That is, was there a large sample size, a cause-and-effect relationship, etc.
If they had more test subjects this would be better science but with only 20 people for the test subjests it is bad science.
4. What did the authors determine about forensic graphology?
The Results achieved were very promising,and an identification accuracy of 96.0% was obtained using the WED classifier. The K-NN classifiergave somewhat poor results. The author believes that Graphology is a good thing to use as evidence and that no two hand writing samples were exaclt ythe same. Graphology can be used to identify a person within a 96.0% accuracy.
5. Name at least five interesting facts from the paper that you think the rest of the class might appreciate.